Discrepancy in the Crack width calc for Euro codes

[Anonymous]

Hello,

I am using 2016 version.

I have been facing a issue for the crack width check as per Euro code. The crack width are changing from 0.0mm to 0.6mm  for increment of 1kNm. It seems some error in the program/I don't know actually. I tried all sort of things but no change. (Similar kind of issue I had faced for AASHTO code as well for stress calc).

Please find the attached screen shot images of design file (I could not able to upload the .sam file extension).

The section is uncracked for 1996kNm and the section is cracked with crack width of 0.602mm for 1997kNm. Please check the file let me know where it went wrong. Since this is relates to reliability of the software.  

At the moment I am working on crucial time schedule to complete the project, these things pulling me down to manual check.  

Early response is much appreciated 

Thanks 

Vijay

[simon_hooper]

Hi Vijay

I just put in your section into the latest version (2018) and the results for 1996 and 1997 kN.m pretty much match....with a crack width of 1.52mm.

Are you able to try with the latest version?

Thanks

Simon

[Anonymous]

Hi Simon,

Thanks for your response. I had checked with 2017 Version but the results were wrong as version 2016. I will need to update 2018 version and check. Meantime I had raised a query ticket to Autodesk from our IT helpdesk. I yet to get the response from them. I would like to know the reason for this discrepancy! I am getting a thought that, at what extent we can rely on this software ? 

Thanks 

Vijay

[simon_hooper]

Hi Vijay

I hope you will get a resolution from the support team.  In the meantime if you could zip up the .sam file and attach I will double check the result in 2017 and 2018 (in case I didn't input exactly the same as in your screenshots).

Thanks

Simon

[Anonymous]

Hi Simon,

Thanks for your reply.

please find the attached .sam file

Thanks 

Vijay

[simon_hooper]

Thanks for the file.

I reproduced the problem with your file in 2017.  I have attached the section I created from scratch which doesn't have the issue in 2017.  I will work on tracing the difference between the 2 files and let you know.

Simon

[simon_hooper]

Hi Vijay

There is something in your file that's not right, but I can't put my finger on it.  I have found if you load your file (in 2017) and go to "General Stress-Strain" and then OK, and back to "Bending, Axial and Shear" the problem is resolved. The file can be saved and the problem doesn't seem to come back.

Simon

[simon_hooper]

It seems that the problem lies with selecting "long term" for the elastic modulus used in the serviceability data.  Ignore my last post, since switching to "General Stress-Strain" and back resets the serviceability calculations to Short term, which is why the problem went away. 

Simon

[Anonymous]

Hi Simon,

Thanks for your response. Yes, it does give proper results short term concrete modulus when we switch back to general stress-strain to bending Axial-shear. But still for Long term the problem persist. What is your advice/solution this issue/discrepancy?   

Thanks 

Vijay

[simon_hooper · Accepted solution]

I have logged this as a defect for further investigation by the software development team. 

There isn't a workaround.  We can't just change the elastic modulus for the concrete property to be the long term value, as this would effect the crack width calculations which need the short term modulus. Also the load duration factor would be different.  But changing the elastic mod to the long term value would give you the stresses and strains which you could then use to adjust the crack width calcs manually.

Simon

[Anonymous]

Hi Simon,

Thank you very much for your actions. 

I had also faced similar issues/situations for serviceability check as per AASHTO LRFT couple of month back and later I had used hand calculations. Would be much appreciated if you add this discrepancy to the development team. 

Many Thanks

Vijay

[Anonymous]

Hello Simon,

After quite long time, this is an follow up to crack width calc as per Euro code. 

I am trying very hard to find update to the Structural Bridge Design software! But not succeeded yet. Can you please help me to get the update for what we found defects for the crack width calc if any ? 

Many Thanks

Vijay

[Pawel.Pulak]

Hello Vijay,

as written by Simon in this post the latest update to the Autodesk Structural Bridge Design 2018, ASBD 2018 Service Pack 1, is now available and can be installed using Autodesk Desktop App.

It is also possible to download it from Autodesk Account using Management > All Products & Services > Structural Bridge Design > Downloads > Updates & Add-ons:

Unfortunately the issue encountered by you is not fixed in it.

Regards,

[Anonymous]

Hello Pawel,

Thanks for the response. I went through the link and update features list found it was not fixed. However, I had raised IT request for the same to update from my side. Can you please tell me when this issue will be fixed and the update made available to us?

Thanks 

Vijay

[Pawel.Pulak]

Hello Vijay,

apologies but I am not able to provide this information yet.

Regards,

[karoly_tarcsi]

Hi Vijay,

Sorry for the late response. I am trying to find an explanation for your problem:

1) According to EN 1992-1-1:2004 Article 7.1(2): cross-sections are assumed to be un-cracked if tensile stress does not exceed f_ct,eff = f_ctm.  See attached ScrrenShot_1.png;

2) Tensile strength is set to f_ctm = -3.5088MPa ( default for C40 concrete) in your model. See ScreenShot_2.png;

3) For a more pronounced effect, I used slightly modified loads: SL3=1996.744kNm and SL4 = 1996.743 during my investigation instead of 1996kNm and 1997kNm as you did;

4) For SL4 load case the tensile strength of the concrete at the extreme bottom fiber is not yet exhausted, therefore there is no crack width calculated. Section is assumed to be un-cracked according to EC 1992-1-1 7.1(2). For this load the bottom tensile stress is just below tensile strength. See left side of ScreenShot_3.  

On the other hand, for SL3 the extreme fiber stress exceeds the tensile strength of the concrete, therefore the section is treated as cracked and 0.602mm is given as crack width. See right side of ScreenShot_3.

5) In real life the cross-section will not have a step behavior change because of a small load increase. Micro cracks will start appearing as the section reaches its tensile limit and these cracks will gradually open up.

Note:

- Eurocode calculates the maximum crack width, so there would be a number of smaller cracks too,

- ASBD seeks for the reinforcing bar with the largest crack width.

Let me know what you think.

Regards

Karoly

[dave_geeves]

I fully concur with the explanation given by Karoly.Tarksi on this problem. The non-linear behavior of the section as the applied moment increases just above the cracking moment is as I would expect.  In an extreme case, consider a stick of chalk with an applied moment that creates an extreme fibre tensile stress just below the tensile limit.  if this moment is increased very slightly then the crack width suddenly goes from 0.0 to infinity and the stick of chalk snaps.  Even if we put a very small amount of reinforcing wire in the stick of chalk, the crack would suddenly appear and the stress distributions would be adjusted so that the loss of tensile force from the cracked chalk would be balanced  by the increase in the steel wire force, where the added strains to achieve this would be large and hence the cracks would be large.  I hope this helps

Kind Regards

Dave Geeves

[Anonymous]

Hello,

May I ask why (in Structural Bridge Design 2018) fctm and fct,eff are given as negative values?

I don't see why values of strength are given as negative values.

Moreover, these figures are positive in Structural Bridge Design 2017. As a result of this discrepancy, the 2 different versions give very different crack widths. For a given section and loading, 2018 gives 0.412mm and 2017 gives 0.127mm. This large discrepancy originates from the fact that fctm and fct,eff are considered as negative values in 2018 (see attached image).

It seems to me that 2018 is incorrect in this respect.

If this could be confirmed.

Thanks,

Omar

[karoly_tarcsi]

Hi Omar,

Thanks for posting.

You appear to use an older version of the software (2018.0?). This bug has been fixed in the 2018.1 release.

Refer to Release Notes 2018.1 section "Design Beams", 5th bullet point: "EU Pre-stressed beam: External exception in SLS-Frequent design when either extreme fiber is in tension and error calculating cracking strain (negative fct,eff)".

Note: ASBD represents tensile stresses/limits as negative values but applies the absolute values where it is required. (like in the case of Exp. 7.9) 

Can you confirm the version of your ASBD? If it is 2018.1 or 2018.2 than can you attach your cross-section for inspection?

Regards,